Move some solvers to optional deps (#112)

* Move some solvers to optional deps

Author: tmigot

Project.toml

@@ -5,48 +5,34 @@ version = "0.1.0"
 
 [deps]
 ADNLPModels = "54578032-b7ea-4c30-94aa-7cbd1cce6c9a"
-CaNNOLeS = "5a1c9e79-9c58-5ec0-afc4-3298fdea2875"
-DCISolver = "bee2e536-65f6-11e9-3844-e5bb4c9c55c9"
 DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"
-FletcherPenaltySolver = "e59f0261-166d-4fee-8bf3-5e50457de5db"
 JSOSolvers = "10dff2fc-5484-5881-a0e0-c90441020f8a"
 JuMP = "4076af6c-e467-56ae-b986-b466b2749572"
-KNITRO = "67920dd8-b58e-52a8-8622-53c4cffbe346"
 LLSModels = "39f5bc3e-5160-4bf8-ac48-504fd2534d24"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 LinearOperators = "5c8ed15e-5a4c-59e4-a42b-c7e8811fb125"
 Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"
 NLPModels = "a4795742-8479-5a88-8948-cc11e1c8c1a6"
-NLPModelsIpopt = "f4238b75-b362-5c4c-b852-0801c9a21d71"
 NLPModelsJuMP = "792afdf1-32c1-5681-94e0-d7bf7a5df49e"
-NLPModelsKnitro = "bec4dd0d-7755-52d5-9a02-22f0ffc7efcb"
 NLPModelsModifiers = "e01155f1-5c6f-4375-a9d8-616dd036575f"
 Percival = "01435c0c-c90d-11e9-3788-63660f8fbccc"
 QuadraticModels = "f468eda6-eac5-11e8-05a5-ff9e497bcd19"
-RipQP = "1e40b3f8-35eb-4cd8-8edd-3e515bb9de08"
-SolverBenchmark = "581a75fa-a23a-52d0-a590-d6201de2218a"
+Requires = "ae029012-a4dd-5104-9daa-d747884805df"
 SolverCore = "ff4d7338-4cf1-434d-91df-b86cb86fb843"
 SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
 
 [compat]
 ADNLPModels = "0.4, 0.5, 0.6, 0.7"
-CaNNOLeS = "0.7"
-DCISolver = "0.4"
 DataFrames = "1"
-FletcherPenaltySolver = "0.2"
 JSOSolvers = "0.10, 0.11"
 JuMP = "1"
-KNITRO = "0.13.0"
 LLSModels = "0.3.6"
 LinearOperators = "2"
 NLPModels = "0.19, 0.20"
-NLPModelsIpopt = "0.10"
 NLPModelsJuMP = "0.11, 0.12"
-NLPModelsKnitro = "0.7"
 NLPModelsModifiers = "0.6.2"
 Percival = "0.6, 0.7"
 QuadraticModels = "0.9.2"
-RipQP = "0.5, 0.6"
-SolverBenchmark = "0.5"
+Requires = "1"
 SolverCore = "0.3"
 julia = "^1.6.0"

docs/Project.toml

@@ -1,17 +1,20 @@
 [deps]
 ADNLPModels = "54578032-b7ea-4c30-94aa-7cbd1cce6c9a"
 CaNNOLeS = "5a1c9e79-9c58-5ec0-afc4-3298fdea2875"
+DCISolver = "bee2e536-65f6-11e9-3844-e5bb4c9c55c9"
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
 GR = "28b8d3ca-fb5f-59d9-8090-bfdbd6d07a71"
 JSOSolvers = "10dff2fc-5484-5881-a0e0-c90441020f8a"
 JSOSuite = "ed6ae0be-a024-11e9-2788-05dbf8cd15d9"
 JuMP = "4076af6c-e467-56ae-b986-b466b2749572"
 Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"
 NLPModels = "a4795742-8479-5a88-8948-cc11e1c8c1a6"
+NLPModelsIpopt = "f4238b75-b362-5c4c-b852-0801c9a21d71"
 NLPModelsJuMP = "792afdf1-32c1-5681-94e0-d7bf7a5df49e"
 OptimizationProblems = "5049e819-d29b-5fba-b941-0eee7e64c1c6"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 QuadraticModels = "f468eda6-eac5-11e8-05a5-ff9e497bcd19"
+RipQP = "1e40b3f8-35eb-4cd8-8edd-3e515bb9de08"
 SolverBenchmark = "581a75fa-a23a-52d0-a590-d6201de2218a"
 SolverCore = "ff4d7338-4cf1-434d-91df-b86cb86fb843"
 

docs/src/benchmark.md

@@ -2,8 +2,7 @@
 
 Benchmarking is very important when researching new algorithms or selecting the most approriate ones.
 
-The package [`SolverBenchmark`](https://github.com/JuliaSmoothOptimizers/SolverBenchmark.jl) exports the function [`bmark_solvers`](https://github.com/JuliaSmoothOptimizers/SolverBenchmark.jl/blob/main/src/bmark_solvers.jl) that runs a set of solvers on a set of problems. `JSOSuite.jl` specialize this function, see [`bmark_solvers`](@ref).
-
+The package [`SolverBenchmark`](https://github.com/JuliaSmoothOptimizers/SolverBenchmark.jl) exports the function [`bmark_solvers`](https://github.com/JuliaSmoothOptimizers/SolverBenchmark.jl/blob/main/src/bmark_solvers.jl) that runs a set of solvers on a set of problems. `JSOSuite.jl` specialize this function, see `bmark_solvers`.
 
 The [JuliaSmoothOptimizers organization](https://juliasmoothoptimizers.github.io) contains several packages of test problems ready to use for benchmarking. The main ones are
 - [`OptimizationProblems.jl`](https://github.com/JuliaSmoothOptimizers/OptimizationProblems.jl): This package provides a collection of optimization problems in JuMP and ADNLPModels syntax;
@@ -51,9 +50,10 @@ For the purpose of this example, we will consider 3 solvers.
 select = ["IPOPT", "TRUNK", "LBFGS"]
 ```
 
-Once the problems and solvers are chosen, the function [`bmark_solvers`](@ref) runs the benchmark. 
+Once the problems and solvers are chosen, the function `bmark_solvers` runs the benchmark. 
 
 ```@example op
+using SolverBenchmark
 stats = bmark_solvers(ad_problems, select, atol = 1e-3, max_time = 10.0, verbose = 0)
 ```
 

docs/src/nls.md

@@ -57,6 +57,7 @@ By default, `JSOSuite.solve` will use a solver tailored for nonlineat least squa
 Nevertheless, it is also possible to specify the solver to be used.
 
 ```@example ex1
+using NLPModelsIpopt
 stats = solve("IPOPT", F, x0, nres, c, l, l)
 ```
 

docs/src/qp.md

@@ -55,5 +55,6 @@ using QuadraticModels, JSOSuite
 As usual, it is also possible to select manually the solver to be used.
 
 ```@example ex1
+  using RipQP
   stats = solve("RipQP", c, H, A, lcon, ucon, name = "eqconqp_QP")
 ```

src/JSOSuite.jl

@@ -1,18 +1,14 @@
 module JSOSuite
 
 # other dependencies
-using DataFrames, JuMP, KNITRO
+using DataFrames, JuMP, Requires
 # stdlib
 using LinearAlgebra, Logging, SparseArrays
 # JSO
 using ADNLPModels, LLSModels, NLPModels, NLPModelsJuMP, QuadraticModels
 using LinearOperators, NLPModelsModifiers, SolverCore
 # JSO solvers
-using CaNNOLeS,
-  DCISolver, FletcherPenaltySolver, JSOSolvers, NLPModelsIpopt, JSOSolvers, Percival, RipQP
-if KNITRO.has_knitro()
-  using NLPModelsKnitro
-end
+using JSOSolvers, Percival
 
 """
     solvers
@@ -58,7 +54,7 @@ push!(
     :KnitroSolver,
     "NLPModelsKnitro.jl",
     :knitro,
-    KNITRO.has_knitro(),
+    false,
     true,
     true,
     true,
@@ -191,7 +187,7 @@ push!(
     :CaNNOLeSSolver,
     "CaNNOLeS.jl",
     :cannoles,
-    true,
+    false,
     false,
     true,
     false,
@@ -210,7 +206,7 @@ push!(
     :IpoptSolver,
     "NLPModelsIpopt.jl",
     :ipopt,
-    true,
+    false,
     true,
     true,
     true,
@@ -229,7 +225,7 @@ push!(
     :DCIWorkspace,
     "DCISolver.jl",
     :dci,
-    true,
+    false,
     false,
     true,
     false,
@@ -248,7 +244,7 @@ push!(
     :FPSSSolver,
     "FletcherPenaltySolver.jl",
     :fps_solve,
-    true,
+    false,
     false,
     true,
     false,
@@ -286,7 +282,7 @@ push!(
     :RipQPSolver,
     "RipQP.jl",
     :ripqp,
-    true,
+    false,
     true,
     true,
     true,
@@ -375,14 +371,16 @@ The list of available solver can be obtained using `JSOSuite.solvers[!, :name]`
 
 ```jldoctest; output = false
 using JSOSuite
-stats = solve("DCISolver", x -> 100 * (x[2] - x[1]^2)^2 + (x[1] - 1)^2, [-1.2; 1.0], verbose = 0)
+stats = solve("TRON", x -> 100 * (x[2] - x[1]^2)^2 + (x[1] - 1)^2, [-1.2; 1.0], verbose = 0)
 stats
 
 # output
 
 "Execution stats: first-order stationary"
 ```
 
+Some solvers are available after loading only.
+
 ```jldoctest; output = false
 using JSOSuite
 # We solve here a quadratic problem with bound-constraints
@@ -420,9 +418,58 @@ include("solve_model.jl")
 
 include("solve.jl")
 
-using SolverBenchmark
+@init begin
+  @require CaNNOLeS = "5a1c9e79-9c58-5ec0-afc4-3298fdea2875" begin
+    JSOSuite.solvers[JSOSuite.solvers.name .== "CaNNOLeS", :is_available] .= 1
+    function solve(::Val{:CaNNOLeS}, nlp; kwargs...)
+      return CaNNOLeS.cannoles(nlp; linsolve = :ldlfactorizations, kwargs...)
+    end
 
-export bmark_solvers
+  end
+end
+
+@init begin
+  @require DCISolver = "bee2e536-65f6-11e9-3844-e5bb4c9c55c9" begin
+    JSOSuite.solvers[JSOSuite.solvers.name .== "DCISolver", :is_available] .= 1
+    function solve(::Val{:DCISolver}, nlp; kwargs...)
+      return DCISolver.dci(nlp; kwargs...)
+    end
+  end
+end
+
+@init begin
+  @require FletcherPenaltySolver = "e59f0261-166d-4fee-8bf3-5e50457de5db" begin
+    JSOSuite.solvers[JSOSuite.solvers.name .== "FletcherPenaltySolver", :is_available] .= 1
+    function solve(::Val{:FletcherPenaltySolver}, nlp; kwargs...)
+      return FletcherPenaltySolver.fps_solve(nlp; kwargs...)
+    end
+  end
+end
+
+@init begin
+  @require NLPModelsIpopt = "f4238b75-b362-5c4c-b852-0801c9a21d71" begin
+    JSOSuite.solvers[JSOSuite.solvers.name .== "IPOPT", :is_available] .= 1
+    include("solvers/ipopt_solve.jl")
+  end
+end
+
+@init begin
+  @require NLPModelsKnitro = "bec4dd0d-7755-52d5-9a02-22f0ffc7efcb" begin
+    @init begin
+      @require NLPModelsKnitro = "bec4dd0d-7755-52d5-9a02-22f0ffc7efcb" begin
+        JSOSuite.solvers[JSOSuite.solvers.name .== "KNITRO", :is_available] .= KNITRO.has_knitro()
+      end
+    end
+    include("solvers/knitro_solve.jl")
+  end
+end
+
+@init begin
+  @require RipQP = "1e40b3f8-35eb-4cd8-8edd-3e515bb9de08" begin
+    JSOSuite.solvers[JSOSuite.solvers.name .== "RipQP", :is_available] .= 1
+    include("solvers/ripqp_solve.jl")
+  end
+end
 
 """
     bmark_solvers(problems, solver_names::Vector{String}; kwargs...)
@@ -454,7 +501,7 @@ All the remaining keyword arguments are passed to the function `SolverBenchmark.
 # Examples
 
 ```jldoctest; output = false
-using ADNLPModels, JSOSuite
+using ADNLPModels, JSOSuite, SolverBenchmark
 nlps = (
   ADNLPModel(x -> 100 * (x[2] - x[1]^2)^2 + (x[1] - 1)^2, [-1.2; 1.0]),
   ADNLPModel(x -> 4 * (x[2] - x[1]^2)^2 + (x[1] - 1)^2, [-1.2; 1.0]),
@@ -480,7 +527,7 @@ KeySet for a Dict{Symbol, DataFrames.DataFrame} with 2 entries. Keys:
 The second example shows how to add you own solver to the benchmark.
 
 ```jldoctest; output = false
-using ADNLPModels, JSOSolvers, JSOSuite, Logging
+using ADNLPModels, JSOSolvers, JSOSuite, Logging, SolverBenchmark
 nlps = (
   ADNLPModel(x -> 100 * (x[2] - x[1]^2)^2 + (x[1] - 1)^2, [-1.2; 1.0]),
   ADNLPModel(x -> 4 * (x[2] - x[1]^2)^2 + (x[1] - 1)^2, [-1.2; 1.0]),
@@ -510,31 +557,39 @@ KeySet for a Dict{Symbol, DataFrames.DataFrame} with 3 entries. Keys:
 
 ```
 """
-function SolverBenchmark.bmark_solvers(
-  problems,
-  solver_names::Vector{String},
-  solvers::Dict{Symbol, Function} = Dict{Symbol, Function}();
-  atol::Real = √eps(),
-  rtol::Real = √eps(),
-  verbose::Integer = 0,
-  max_time::Float64 = 300.0,
-  max_eval::Integer = 10000,
-  max_iter::Integer = 10000,
-  kwargs...,
-)
-  for s in solver_names
-    solvers[Symbol(s)] =
-      nlp -> solve(
-        s,
-        nlp;
-        atol = atol,
-        rtol = rtol,
-        verbose = verbose,
-        max_time = max_time,
-        max_eval = max_eval,
-      )
+function bmark_solvers end
+
+@init begin
+  @require SolverBenchmark = "581a75fa-a23a-52d0-a590-d6201de2218a" begin
+
+    function SolverBenchmark.bmark_solvers(
+      problems,
+      solver_names::Vector{String},
+      solvers::Dict{Symbol, Function} = Dict{Symbol, Function}();
+      atol::Real = √eps(),
+      rtol::Real = √eps(),
+      verbose::Integer = 0,
+      max_time::Float64 = 300.0,
+      max_eval::Integer = 10000,
+      max_iter::Integer = 10000,
+      kwargs...,
+    )
+      for s in solver_names
+        solvers[Symbol(s)] =
+          nlp -> solve(
+            s,
+            nlp;
+            atol = atol,
+            rtol = rtol,
+            verbose = verbose,
+            max_time = max_time,
+            max_eval = max_eval,
+          )
+      end
+      return SolverBenchmark.bmark_solvers(solvers, problems; kwargs...)
+    end
+
   end
-  return bmark_solvers(solvers, problems; kwargs...)
 end
 
 """